http://www.diva-portal.org

This is the published version of a paper published in BMC Genomics.

Citation for the original published paper (version of record):

Niemiec, M J., Grumaz, C., Ermert, D., Desel, C., Shankar, M. et al. (2017)Dual transcriptome of the immediate neutrophil and Candida albicans interplay.BMC Genomics, 18: 696https://doi.org/10.1186/s12864-017-4097-4

Access to the published version may require subscription.

N.B. When citing this work, cite the original published paper.

Permanent link to this version:http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-102836

RESEARCH Open Access

Dual transcriptome of the immediateneutrophil and Candida albicans interplayMaria J. Niemiec1,9, Christian Grumaz2, David Ermert1,10, Christiane Desel3,11, Madhu Shankar1, José Pedro Lopes1,Ian G. Mills4,5,6, Philip Stevens7,8, Kai Sohn2 and Constantin F. Urban1*

Abstract

Background: Neutrophils are traditionally considered transcriptionally inactive. Compared to other immune cells,little is known about their transcriptional profile during interaction with pathogens.

Methods: We analyzed the meta-transcriptome of the neutrophil-Candida albicans interplay and the transcriptomeof C. albicans challenged with neutrophil extracellular traps (NETs) by RNA-Seq, considering yeast and hyphaindividually in each approach.

Results: The neutrophil response to C. albicans yeast and hyphae was dominated by a morphotype-independentcore response. However, 11 % of all differentially expressed genes were regulated in a specific manner whenneutrophils encountered the hyphal form of C. albicans. While involving genes for transcriptional regulators,receptors, and cytokines, the neutrophil core response lacked typical antimicrobial effectors genes. Genes of theNOD-like receptor pathway, including NLRP3, were enriched.Neutrophil- and NET-provoked responses in C. albicans differed. At the same time, the Candida transcriptome uponneutrophil encounter and upon NET challenge included genes from various metabolic processes and indicate amutual role of the regulators Tup1p, Efg1p, Hap43p, and Cap1p. Upon challenge with neutrophils and NETs, theoverall Candida response was partially morphotype-specific. Yet again, actual oppositional regulation in yeasts andhyphae was only detected for the arginine metabolism in neutrophil-infecting C. albicans.

Conclusions: Taken together, our study provides a comprehensive and quantitative transcript profile of theneutrophil–C. albicans interaction. By considering the two major appearances of both, neutrophils and C. albicans,our study reveals yet undescribed insights into this medically relevant encounter. Hence, our findings will facilitatefuture research and potentially inspire novel therapy developments.

Keywords: Candida, Neutrophils, Dual transcriptome, Extracellular traps, NOD-like receptor pathway, Nutritionalimmunity, Morphotype

BackgroundNeutrophils are essential to fight off microbial infectionsusing a well-orchestrated set of intra- and extracellularantimicrobial processes [1]. These phagocytes mature inthe bone marrow and are constantly released into theblood stream. In circulation, they are fully matured.Their lifespan is comparably short between one andseveral days [2, 3]. At the site of infection, pathogen-associated molecular patterns (PAMPs) are at first

recognized by pattern recognition receptors (PRRs) onthe surface of neutrophils. Of all fungi-sensing PRRs, ap-proximately 50% are present on neutrophils such asToll-like receptors, C-type lectin receptors, complementreceptor CR3, and Fc receptors [4–6]. Most of thesePRRs recognize fungal cell wall components, like e.g.mannan and β-1,3-glucan [7]. The next step of theneutrophil-microbe interaction is phagocytosis. Phago-somes fuse with cytoplasmic vesicles (granules) to forma highly hostile environment that contains antimicrobialenzymes and peptides as well as reactive oxygen species(ROS) [8]. Within mature phagosomes neutrophilseradicate pathogens efficiently [1, 9, 10]. Neutrophil

* Correspondence: constantin.urban@umu.se1Department of Clinical Microbiology, Umeå Centre for Microbial Research(UCMR) & Laboratory of Molecular Infection Medicine Sweden (MIMS), UmeåUniversity, Umea, SwedenFull list of author information is available at the end of the article

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Niemiec et al. BMC Genomics (2017) 18:696 DOI 10.1186/s12864-017-4097-4

granule content is in addition released into the extracel-lular space to destroy not internalized microbes [11].Neutrophil extracellular traps (NETs) are a second exter-nal killing mechanism [12]. Unlike during degranulation,neutrophils die during NET formation [13]. Before that, acontrolled loss of intracellular membrane integrity allowsthe contents of formerly separated compartments to mix.Therefore NET fibers contain nuclear chromatin decoratedwith granular and cytoplasmic effectors [14–16]. Inaddition to directly affecting the microbial invader, neutro-phils are able to release cytokines in order to recruit or ma-nipulate other immune cells [17, 18].A pathogen frequently encountered by neutrophils is

Candida albicans. Today, Candida ssp. cause most ofthe opportunistic fungal infections worldwide. Neverthe-less, C. albicans is a commensal in more than 50% of thehuman population [19]. It colonizes different mucosalniches, such as for instance the gastrointestinal (GI)tract, without causing apparent symptoms. However,upon perturbance of immune barriers or functions, C.albicans can turn into a bona fide pathogen [20]. Thefungus causes a wide range of diseases from superficial, in-vasive to systemic infections. Invasive forms of candidiasisare often correlated to extremes of age, organ transplant-ation, blood cancer and HIV/AIDS. Epidemiological stud-ies show that Candida infections obtained in intensivecare units can be fatal in up to 50% of the cases [21, 22].C. albicans pathogenicity is based on a variety of fea-

tures, amongst which the ability to change morphotypesis essential. C. albicans can grow as ovoid-shaped, bud-ding yeast or as filamentous hyphae. A considerablenumber of studies have investigated the role of the re-spective morphotypes during in vivo infections. It wasshown that mutants locked in one or the other morpho-type are avirulent [23]. Furthermore, it was revealed thatboth morphotypes can be isolated from infection sites[24]. Human neutrophils are able to eradicate bothmorphotypes by intra- and extracellular mechanisms[15, 25]. Several studies have been performed to eluci-date the antifungal mode of action of NETs. The proteincomplex calprotectin, a heterodimer composed of theproteins S100A8 and S100A9, is a zinc and manganesechelator and was found to be associated with NETs afterNETosis. Among the NET-associated host effectors, cal-protectin is the key effector in NETs towards fungalpathogens [16, 26–29].It is likely due to the short life span and completed

maturation of neutrophils that their transcriptional cap-acity is underestimated resulting in a small number oftranscriptional studies in neutrophils [30]. Early work re-vealed RNA and protein synthesis in neutrophils in the1970ies [31]. In 2004, a microarray-based analysis of theneutrophil transcriptional response was performed usinglive, opsonized Escherichia coli, lipopolysaccharide

(LPS), and formyl-Met-Leu-Phe (fMLF) as stimuli [32].Transcriptional responses of neutrophils and monocytestowards C. albicans were analyzed by microarray con-firming a transcriptional activation of neutrophils by thefungus [33]. Remarkably, the response involved genes act-ing in cell communication, e.g. cytokines, but lacked typ-ical antimicrobial effectors needed for a promptdeactivation of pathogens. More recently, a study com-pared the transcriptional effects of priming agents TNF-αand GM-CSF on human neutrophils using RNA-Seq [34].Many studies have analyzed the gene expression in C.albicans reacting to a variety of stimuli. Biotic stimuli wereamongst others full blood [35], granulocytes isolated fromblood [36] and macrophages [37, 38].The present study analyzed the transcriptional re-

sponse of human neutrophils and C. albicans duringtheir interaction in greater detail by employing RNA-Seq. First, we performed in vitro infections of neutro-phils by C. albicans yeast or C. albicans hyphae andinvestigated the early transcriptional response from bothinteraction partners. Candida yeast and hyphae are bothsusceptible towards NETs [15]. However, it is incom-pletely understood how C. albicans transcriptionally re-sponds to NET-mediated stress. Thus, we challenged C.albicans yeasts and hyphae with NETs and analyzedtheir transcriptional response.Among the investigated time points we found the stron-

gest transcriptional response of neutrophils to C. albicansafter 60 min. Genes induced in neutrophils code for pro-teins involved in transcriptional regulation, gene expres-sion, cytokine production, stress responses, and apoptosisdelay. The majority of 318 differentially expressed genes(DEGs) was up-regulated. Notably, we found approxi-mately 11% of the DEGs to be morphotype-specificallyregulated upon encounter of C. albicans hyphae. Incontrast, the transcriptional response of C. albicans wasspread more evenly over the entire time course analyzed.Overall, C. albicans’ response towards neutrophils wasdominated by metabolic genes which are controlled by thetranscriptional regulators Tup1p, Efg1p, Hap43p, andCap1p. No apparent morphotype specificity was discov-ered in C. albicans challenged with NETs and overall thetranscriptional response of C. albicans to NETs was verydifferent from the response to intact neutrophils. Ourfindings give detailed insight into the neutrophil–C. albi-cans interaction and contribute to the understanding ofthis exceptional host-pathogen interaction.

ResultsTranscriptional response of neutrophils to C. albicansoccurs predominantly after 60 min and is dominated by amorphotype-independent core responseSince C. albicans is a polymorphic pathogen and mecha-nisms applied by neutrophils to discriminate between

Niemiec et al. BMC Genomics (2017) 18:696 Page 2 of 21

the C. albicans morphotypes remain incompletelyunderstood [39–41], we analyzed transcriptional changesin neutrophils isolated from healthy donors upon yeastand hypha infection separately. DEGs were determinedthrough comparison of infected and uninfected neutro-phils. We have shown earlier that quantification basedon dry mass was preferable to determine comparableamounts of C. albicans yeasts and hyphae [39]. In thepresent study we infected with similar dry masses, whichcorrelated well with surface area of yeast and hyphalcells resulting in comparable infectious loads and con-tact sites for neutrophils. The infection of neutrophilswas analyzed after 15, 30 and 60 min to represent differ-ent stages of in vitro infection: recognition and adhesion,early phagocytosis, and entire phagocytosis of yeast orwrapping of hyphae, respectively. For both infections,neutrophil DEGs were most abundant at 60 min post in-fection (p.i.) indicating that the immediate attack of neu-trophils does not majorly require de novo synthesis ofeffector molecules (Fig. 1a and b, yeasts and hyphae, re-spectively). Of note, the purity of our neutrophil prepa-rations was verified by flow cytometry (Additional file 1:Figure S1). CD11b-positive and MHC class II-negativecells were considered as neutrophilic granulocytes. Theirshare was consistently higher than 93%. CD3-positive Tcells and CD19-positive B cells accounted for less than1% of the purified cells. The majority of cells carried-over were eosinophilic granulocytes, showing character-istic high side scatter and auto-fluorescence properties,whereas CD14-positive monocytes constituted below0.1% of the cell suspension.For the neutrophil response, an average of 35.4 million

reads was sequenced. Of those, 25.9 million reads were

uniquely mapped (73%) and therefore could be utilizedfor subsequent gene expression analyses Additional file 2:Table ST1).As an overall trend, the transcriptional response of neu-

trophils peaked after 60 min, independent of the Candidamorphotype used for the infection (Fig. 1a and b,Additional file 3: Figure S2A, Additional file 4: TableST2).In detail, slight differences could be detected: Nearly allDEGs in yeast-infected neutrophils were exclusivelyaffected at 60 min p.i. and no DEGs were found to beconstitutively affected throughout the entire infection(Fig. 1a). In contrast, a group of 16 DEGs, all up-regulated, appeared at all times in the hypha-infected neu-trophils (Fig. 1b).When comparing all the neutrophil genes up- and

down-regulated in response to either yeasts or hyphae,the majority belonged to a 225 gene-containingmorphotype-independent ‘core neutrophil response’. Ofthose, 218 were up-regulated and only 7 down-regulated(Fig. 1c and d). When grouping the 225 genes in 9 majorclasses according to function, the relative abundance in-dicated a high level of cellular reprogramming, inductionof gene expression, cytokine production and stress re-sponse (Table 1).In addition to function, the union set of all neutrophil

DEGs, which comprise 318 genes, were also clustered byexpression pattern (Additional file 5: Figure S3). Eightdifferent clusters were identified by QT-Clustering usingMayday and most DEGs were assigned to cluster 8 con-taining genes with increasing expression from 15 min to60 min (Additional file 5: Figure S3: cluster 8) [42]. Inthree clusters, we identified 36 of the DEGs to bemorphotype-specifically regulated in hypha-infected

Fig. 1 Overlaps of DEGs in neutrophils infected with C. albicans yeast and hyphae. Overlap of DEGs in neutrophils infected with C. albicans yeast(a) and hyphae (b) over time. Overlap of morphotype-specific neutrophil response of induced (c) and repressed (d) DEGs. Samples from twoindependent experiments using different blood donors were analyzed, n = 2

Niemiec et al. BMC Genomics (2017) 18:696 Page 3 of 21

neutrophils (Additional file 5: Figure S3: cluster 1,3 + 7). Those genes code for cytokines, receptors, poten-tial PRRs, and transcription regulators. As these are pro-teins of versatile function and a noticeable proportion ofthese specifically regulated genes has not been assignedto a particular function in neutrophils yet, their abun-dance does not point towards one clear key mode of ac-tion of neutrophils battling C. albicans hyphae.In summary, both quantity and intensity of the tran-

scriptional response of human neutrophils towards C.albicans yeasts and hyphae increased from 15 min to60 min p.i.. This indicates that different stages of inter-play might be implied. The immediate early response ofneutrophils to Candida involves minor but distinct tran-scriptional activity, but might be executed mostly byposttranslational events. In contrast, later responsesseem to require de novo synthesis of proteins. A concisebut noticeable proportion of the neutrophil DEGs wasregulated in a morphotype-specific manner upon en-counter of C. albicans hyphae.

Expression of transcription regulators, PRRs and cytokinereceptors promotes inflammatory responses and a switchfrom migration to battleThe group of genes involved in transcriptional regulationcontained 49 genes (Fig. 2a) clustered by their dynamicexpression patterns (Additional file 5: Figure S3). Themajority of differentially expressed regulator genesbelonged to cluster 8 with increasing transcript levelsover time. Many of these regulator genes were zinc-finger repressors or leucine zipper motif-containing fac-tors. These regulators have been described to play a roleduring inflammation and oxidative stress responses. Re-markably, genes of all members of the FOS family (FOS,FOSB, FOSL1, FOSL2) were differentially expressed. Thecoded proteins of FOSL2 and FOSB dimerize with JUN

to form the transcription factor complex AP-1 [43].Interestingly, EGR1, even though considered as unclus-tered, showed highest transcript levels after 30 min ofinfection, therefore being the most induced DEG at thistime point (Additional file 6: Table ST3). A role for tran-scriptional activator EGR1 during inflammatory re-sponses of neutrophils has been described [44].Further, we revealed several transcription regulator

genes to be expressed at first and repressed at later timepoints, such as DPF2 (Additional file 5: Figure S3, clus-ter 6). The coded protein is crucial for induction ofapoptosis upon e.g. starvation. Interestingly, we identi-fied a higher relative abundance of transcription regula-tor genes among the most affected neutrophil genesduring yeast infection compared to hypha infection(Additional file 6: Table ST3).The group of receptor genes we found to be differentially

expressed in neutrophils consists of 22 genes (Fig. 2b). Themajority of these genes was down-regulated at first butsubsequently induced to statistically significant levels at60 min p.i. (Additional file 5: Figure S3, cluster 8). For in-stance, IL1R1 codes for the IL-1 receptor type I and is pro-inflammatory. C3AR1 and C5AR1 are the genes of thecomplement component 3a receptor 1 and complementcomponent 5a receptor [45]. Interestingly, we identifiedthe gene TREM1 and NLRP3 to be up-regulated. TREM-1is a pro-inflammatory receptor that interacts with theTLR4-mediated response [46]. NLRP3 encodes for Nalp3(or NLRP3); an intracellular PRR capable of triggeringinflammasome activation [47].In summary, the group of differentially expressed tran-

scriptional regulators indicates induction of an inflam-matory and oxidative stress response and simultaneousrepression of apoptosis, whereas the regulation patternof neutrophil receptor genes indicates that neutrophilsprepare to phase out migration and engage in directinteraction with the fungal pathogen.

Major immunological pathways and cytoskeleton areaffected in Candida-infected neutrophilsThe biggest group of DEGs in Candida-infected neutro-phils were genes whose products function in cell signaling.The group consisted of 71 genes (Fig. 2c) and a number ofsignaling pathways were affected: NfκB, MAPK, JAK-STAT,actin dynamics and membrane trafficking, complement sys-tem, cell cycle, and apoptosis were most evident.We identified 8 DEGs related to the NfκB pathway

and the same number of genes involved in the MAPKpathway. The majority of these 16 genes showed in-creased transcript levels at 60 min p.i. (Additional file 5:Figure S3, cluster 8). NfκB-related genes were for in-stance IRAK2 coding for IL-1 receptor-associated kinase2 and TICAM1 coding for the toll-like receptor mol-ecule adaptor 1 used by TLR3 (not present on

Table 1 Transcriptional re-shaping of the neutrophil duringC. albicans infection

Cellular function %

Transcription factors and regulators 15.6

Gene expression 11.5

Receptors 6.4

Signaling 22.0

Cytokines 6.4

Adhesion and migration 2.8

Membrane trafficking & phagocytosis 6.4

Metabolism, stress response, transport 17.4

Uncharacterized 11.5

Entity of neutrophil DEGs affected during infection with C. albicans yeast andhyphae were assigned to 9 different cellular functions and the relativeabundance of DEGs in this group was determined. Assignment to functionwas done using NCBI, KEGG and GeneCards® databases

Niemiec et al. BMC Genomics (2017) 18:696 Page 4 of 21

neutrophils) and TLR4. DEGs involved in the MAPKpathway were for example MAP3K8 and MAPKAPK2.The products of both genes are mediators in TLRsignaling.A group of 30 DEGs codes for signaling proteins in-

volved in actin dynamics and membrane trafficking mostprobably representing cytoskeleton rearrangements dueto direct interaction with fungal cells (Fig. 2c). Amongthose we found 6 small GTPase genes associated to ves-icular trafficking, as well as genes coding for proteins ofthe Ras superfamily of GTPases and Rho GTPase acti-vating or binding proteins that play a role in F-actin andpseudopodia formation.In addition, we found 7 apoptosis-related genes clus-

tered together that were exclusively statistically signifi-cantly up-regulated at 60 min p.i. (Additional file 5:Figure S3, cluster 8). Of those, several are anti-apoptotic,such as the gene BFAR, an apoptosis regulator with anti-apoptotic activity.

In summary, the pattern of induction and repressionof signaling genes in C. albicans-infected neutrophilsrevealed the involvement of major immunologicalpathways, most prominently NfκB and MAPK. This cor-responds to the receptors and regulators also found tobe affected, such as IL1-, complement- and TLR-mediated signaling. The presence of several DEGs in-volved in apoptosis, most of them inhibiting, indicatesdelay of apoptosis. In addition, the rearrangement ofcytoskeleton and vesicular trafficking were reflected bynumerous DEGs.

Enrichment analyses reveal activation of the NOD-likereceptor signaling pathway in Candida-triggeredneutrophilsIn addition to the functional description and analysis byfold-change of expression, we further compared thecomplete transcriptome of C. albicans-stimulated neu-trophils to pathway databases for mammalian cells. The

Fig. 2 Regulation of regulator, receptor, cytokine, and signaling DEGs affected during neutrophil-C. albicans encounter. Neutrophil DEGs from the‘neutrophil core response’ upon C. albicans infection were grouped according to the function of their coded protein. Those coding for transcriptionalregulators (a), receptors (b), signaling effectors (c), and cytokines (d) are plotted. The transcript profile is displayed for 15 min, 30 min and 60 min p.i.with yeast or hyphae. Red indicates up-regulation, blue indicates down-regulation. Samples from two independent experiments using different blooddonors were analyzed, n = 2

Niemiec et al. BMC Genomics (2017) 18:696 Page 5 of 21

318 DEGs identified were analyzed by ENRICHnet,KEGG pathway enrichment, and Gene Trail. As men-tioned earlier, we have identified the gene NLRP3 to beup-regulated in neutrophils. In line with this, the nuclearoligomerization domain (NOD)-like receptor signalingpathway was identified by all three approaches to beenriched. Of a pathway of 60 genes, seven were foundwithin the 318 DEGs. These seven genes are NLRP3,CXCL1, CXCL2, BIRC3, IL1B, IL8, RIPK2. These sevengenes are all up-regulated over the course of hypha andyeast infection over time, reaching their highest ex-pression levels predominantly at the 60-min time point(Fig. 2). NOD and NOD-like receptors are intracellularPRRs that activate an inflammatory response. The NLRPsubfamily is involved in inflammasome formation [48].To gather further information on inflammasome activa-tion, we analyzed caspase-1 activation in neutrophils in-fected with either Candida yeast or hyphae after oneand three h of incubation (Fig. 3). In accordance withtranscriptional data we observed caspase-1 activation byyeast and hypha infection. The latter resulted in strongeractivation of caspase-1 than yeast infection after 1 h(Fig. 3a) and was similar in both after 3 h (Fig. 3b).

Furthermore, we analyzed secretion of IL-1β, one of themajor products of inflammasome activation. After threeand 6 h of infection IL-1β concentration increased in su-pernatants reflecting caspase-1 activation (Fig. 3c and d).This finding supports transcriptional up-regulation ofthe NLRP3 gene and provides further indications that C.albicans might trigger NLRP3-dependent activation ofinflammasome-dependent pyroptosis in neutrophils.

Neutrophils transcribe cytokine genes upon infection butnot genes coding for archetype antimicrobial factorsInterestingly, 16 cytokine genes were differentially regu-lated (Fig. 2d) with non-significant or marginal transcriptlevels shortly after infection but increasing transcript levelsover time (Additional file 5: Figure S3, cluster 5 and 8).The cytokines coded by the up-regulated genes are for in-stance chemotactic to neutrophils, such as CXCL1, CXCL2and IL8. Further, IL1A and IL1B were also up-regulatedwhose products both bind to IL-1 receptors I and II. Neu-trophil cytokine expression thus indicated orchestration ofneutrophil recruitment and initiation of cross-talk withother immune cell types. To verify the role of neutrophilsas cytokine producing cells, we infected freshly isolated

Fig. 3 C. albicans triggers inflammasome activation in neutrophils quantified by Caspase-1 activation. Caspase-1 activity of C. albicans hyphae andyeast after challenging with human neutrophils for 1 h (a) and 3 h (b). As positive control neutrophils were stimulated with LPS (1 μg/ml),followed by ATP (5 mM, 1 h before adding the FLICA reagent). Amounts of IL-1β in supernatants of neutrophils challenged with C. albicans yeastor hyphae were quantified by ELISA after 3 h (c) and 6 h (d) of incubation. Shown are one representative experiment in triplicate for each assayfrom 2 experiments in total with very similar results, n = 2 (3). Statistical analysis was performed using a One-way ANOVA with Dunett’s post-testwith * = p < 0.05, *** = p < 0.001 and **** = p < 0.0001

Niemiec et al. BMC Genomics (2017) 18:696 Page 6 of 21

neutrophils with dead C. albicans yeast and hypha cells for18 h and quantified the resulting secretion of 10 cytokineswith a Bio-Plex® approach from neutrophils of 5 individualdonors (Table 2 and Fig. 4).Among the 5 cytokines secreted in a Candida-

dependent manner, we identified 4 cytokines that were se-creted in higher concentrations in the cocultivation withhyphae than with yeasts: MIP-1β, M-CSF, IL-1β, and IL-8.Consistent with this finding, the RPKM values of the cod-ing genes for these cytokines were slightly higher at60 min p.i. when neutrophils encountered C. albicans hy-phae, but not the yeast form (Table 2). In addition, the se-cretion of TNF-α was induced equally by both Candidamorphotypes, while in this case the RPKM values wereslightly higher in the yeast infection (Table 2). However,we could not detect inducible secretion of Gro-α, IL-18,IL-1α, IL-1ra, and MIF. Similar trends as with deadhyphae were observed when neutrophils were infectedwith live C. albicans yeast that form hyphae during the ex-periment. (Additional file 7: Figure S4). Of note, the differ-ential expression of these 10 cytokine genes until 60 minp.i. was not entirely reflected by the ELISA-based cytokinesecretion assay 18 h p.i. (Table 2). This indicates transcrip-tion of certain cytokine genes, secretion of pre-stored cy-tokines or shorter half-life of other cytokines. The overlapof transcript- and ELISA-based cytokine analyses confirmshigh reproducibility of our methods, despite possibledonor variations of primary neutrophils.In agreement with earlier investigations, most genes

coding for typical neutrophil effector molecules, knownto be formed during granulopoiesis, were not differen-tially expressed under tested conditions [11, 33]. How-ever, analysis of the average transcription level, reflectedby the RPKM value, revealed considerable variation

between different effector molecules (Table 3) rangingfrom nearly no transcripts (e.g. defensin genes), via lowand intermediate number of transcripts (e.g. cathelicidingene CAMP) to higher levels for some genes coding fac-tors related to the NADPH oxidase complex. Among theclassical antimicrobials, the genes coding for the proteincomplex calprotectin (S100A8 and S100A9) stood out.With constitutively high transcript levels showing RPKMvalues of 4721 and 17,890 in average those genes corres-pond to 1.39% and 5.25% of all RPKM values in unin-fected and infected neutrophils, respectively. In otherwords, more than 6.5% of the entire neutrophil produc-tion of transcripts is dedicated to these two genes. Thisis in good agreement with the notion that the S100A8/A9 complex constitutes the major cytoplasmic protein ofneutrophils [29].In summary, we confirmed that mature neutrophils

have seized transcription of most antimicrobial factorgenes. In contrast, numerous cytokines are transcrip-tionally induced during infection. This was additionallyvalidated using protein quantification in supernatants ofC. albicans-infected neutrophils.

C. albicans response to neutrophils is dependent onmorphotype and time pointFor the in vitro infections analyzed later by RNA-Seq,we allowed Candida to adjust to the infection-relatedtemperature and pH before addition of neutrophilsfor 30 min. The transcripts of these adjusted, but un-challenged fungal cells served as reference for DEGidentification for each time point throughout thecourse of the infection. To have a comparable infec-tious load we adjusted the dry mass of Candida asdescribed previously [39]. An average of 13.1 million

Table 2 Cytokine secretion of neutrophils infected with C. albicans yeasts and hyphae

Cytokine Secretion Gene Expression (RPKM) h60’/y60’

FC [log2] DEG

Control 60 min y 60 min h 60 min y 60 min h

MIP-1β UP-h CCL4 54.4 171.8 194.5 1.1 2.20 2.35 yes

IL-1β UP-h IL1B 242.3 717.1 960.8 1.3 2.11 2.50 yes

IL-8 UP-h IL8 9325.5 20,837.7 28,971.6 1.4 1.68 2.13 yes

M-CSF UP-h CSF1 10.6 11.8 13.6 1.2 0.68 0.86 no

TNFα UP TNF 30.6 39.6 30.6 0.8 0.90 0.50 no

Gro-α n.s. CXCL1 336.2 643.0 674.5 1.0 1.47 1.52 yes

IL-1α n.s. IL1A 2.6 33.3 37.7 1.1 4.24 4.39 yes

IL-1 ra n.s. IL1RN 225.0 486.0 524.3 1.1 1.65 1.74 yes

IL-18 n.s. IL18 0.8 0.4 0.4 1.0 −0.34 −0.36 no

MIF n.s. MIF 0.8 0.4 0.3 0.7 −0.41 −0.91 no

Neutrophils were infected with dead C. albicans yeast or hypha cells and secreted cytokines were detected by Bio-Plex® at 18 h p.i.. Cytokines were either inducedsimilarly with both (UP), secreted more in the presence of hyphae (UP-h), or not secreted (n.s.). Cytokine gene transcript levels (by RPKM) were analyzed after60 min of stimulation, transcript levels in the infection with yeast and hyphae were related (h60’/y60’). Cytokine gene fold changes of expression (FC) are displayedafter 60 min of cocultivation of neutrophils with C. albicans yeasts or hyphae. Statistically significant DEGs are pointed out

Niemiec et al. BMC Genomics (2017) 18:696 Page 7 of 21

reads was sequenced. Of those, 10.3 million readswere uniquely mapped (79%) and therefore utilizedfor gene expression analyses (Additional file 2: TableST1). Only those Candida genes were considered asdifferentially expressed which changed transcript levelhigher than 4-fold at 15, 30 and 60 min neutrophilencounter compared to resting Candida cells (Add-itional file 8: Table ST4). Overall, the gene expressionupon neutrophil encounter in C. albicans yeasts wasaltered more strongly than in hyphal cells. At timepoints 15 min, 30 min and 60 min p.i., 258, 383, and624 genes were differentially expressed in yeast cellscompared to the reference, respectively. In hyphae,73, 95, and 83 genes were found to be DEGs

(Additional file 3: Figure S2B). Since our experimentalconditions per se induce the switch from yeast tohypha, the difference between the numbers of DEGsin both morphotypes was not unexpected (Add-itional file 9: Figure S5C + D).In contrast to the response by neutrophils, the overlap

of differentially-regulated Candida genes throughout theentire infection was more abundant, but time pointspecificity could be observed as well (Additional file 3:Figure S2B, Additional file 9: Figure S5A + B). In gen-eral, more genes were involved in the response of C.albicans yeast towards neutrophils, but there is aconsiderable overlap between the yeast and the hyphalresponse (Additional file 9: Figure S5C + D).

Fig. 4 Cytokine secretion by neutrophils upon C. albicans infection. Neutrophils were stimulated either with thimerosal-killed C. albicans yeast,hyphae or left untreated. Supernatants of 18 h co-incubation were analyzed for cytokine release by Bio-Plex®. Each dot represent a value from anindividual neutrophil donor, n = 5. Statistical analysis was performed using a One-way ANOVA with Bonferroni’s post-test with ** = p < 0.01,*** = p < 0.001 and **** = p < 0.0001

Niemiec et al. BMC Genomics (2017) 18:696 Page 8 of 21

For our analysis, we focused on the entity of genes dif-ferentially regulated in C. albicans hyphal cells uponneutrophil challenge (a total of 135 genes) and comparedtheir level of regulation in yeasts and in hyphae duringthe infection. We refer to these 135 genes as the‘Candida core response’ (not to be mistaken with theso-called ‘Candida core stress response’ by [49]). Indoing so, we take into account that the in vitro infectionconditions induce hyphal growth in Candida yeast cellsand aim to focus on the general response of Candida to-wards human neutrophils (Fig. 5). A group of 57 geneswas induced and 59 genes were repressed in both, yeastand hyphae (Fig. 5a and b). All genes described in the

following were allocated to cellular functions using theCandida Genome Database and Kyoto Encyclopedia ofGenes and Genomes - a GO-enrichment analysis wasperformed to identify the metabolic pathways affected inC. albicans upon neutrophil encounter (Fig. 5c).

C. albicans alters primary sugar and amino acidmetabolism upon neutrophil encounter and generegulation is governed by few transcription factorsIn agreement with earlier transcription studies usingblood or phagocytes to trigger Candida yeast, our GOenrichment analysis revealed that the primary metabol-ism of C. albicans is strongly affected during neutrophil

Table 3 Unaffected expression of neutrophil effectors during C. albicans infection

Gene Protein Average RPKM ± SD Rated

S100A9 S100A9 17,890 3030 ++++

S100A8 S100A8 4721 837 ++++

NCF2 NADPH oxidase, p67 669 103 +++

LYZ lysozyme C 435 91 +++

S100A12 S100A12 372 78 +++

NCF4 NADPH oxidase, p40 293 40 +++

MMP9 gelatinase B 258 49 +++

RAC2 NADPH oxidase, Rac1 or Rac2 236 29 +++

CYBA NADPH oxidase, p22-PHOX 115 19 ++

NCF1 NADPH oxidase, p47 62 13 ++

CAT catalase 51 8 ++

RAC1 NADPH oxidase, Rac1 or Rac2 25 5 +

CAMP cathelicidin, LL-37 25 6 +

CYBB NADPH oxidase, gp91-PHOX 16 3 +

HMGB1 high-mobility group protein 1/amphoterin 8 1 +

TCN1 transcobalamin-I 4 1 –

CRISP3 cysteine-rich secretory protein-3 1 0 –

LTF lactotransferrin 1 0 –

ELANE neutrophil elastase 0 0 –

DEFA1B defensin 1 0 0 –

DEFA1 neutrophil defensin 1 0 0 –

MMP8 neutrophil collagenase 0 0 –

CTSG cathepsin G 0 0 –

MPO myeloperoxidase 0 0 –

PRTN3 proteinase 3 0 0 –

BPI bactericidal/permeability-increasing protein 0 0 –

DEFA3 neutrophil defensin 3 0 0 –

DEFA4 neutrophil defensin 4 0 0 –

DEFA5 defensin 5 0 0 –

DEFA6 defensin 6 0 0 –

The transcript levels of typical neutrophil antimicrobial effector genes that were not differentially expressed during cocultivation with C. albicans was determinedover all time points and conditions. Average and SD RPKM values are displayed and rated to illustrate the expression level. ++++: RPKM >1000; +++: RPKM ≥200;++: RPKM ≥30; +: RPKM ≥5; −: RPKM ≥0

Niemiec et al. BMC Genomics (2017) 18:696 Page 9 of 21

encounter [35–38]. We identified acetate catabolism,carboxylic acid catabolism, and oxidation-reduction asthe three commonly induced processes (Fig. 5c). Becauseof the close relationship of these metabolic pathways, wefound genes assigned to more than one. The inducedgenes ICL1 and FDH1 are each part of two of the threepathways. They code for enzymes involved in the glyoxy-late cycle that has been shown to be induced uponphagocytosis in macrophages and in neutrophils [38,50]. Thus, our RNA-Seq transcriptional analysis con-firms relevant C. albicans pathways originally identifiedby microarrays. Among the oxidation-reduction genes

induced, we found genes coding for different oxidore-ductases and for enzymes involved in resistance againstnitric oxide stress. The latter is in good agreement witha recent study revealing that nitric oxide stress resist-ance is essential for survival of C. albicans under neutro-phil attack [50].Similarly, four processes were revealed to be

commonly repressed in neutrophil-triggered Candidacells: glycolysis, energy reserve metabolism, mono-carboxylic acid metabolism, and oxidation-reduction(Fig. 5c). For instance, 10 genes of enzyme assignedto glycolysis were down-regulated. Thus, genes for

Fig. 5 Core response of C. albicans during neutrophil infection. Candida genes differentially expressed in hyphae upon neutrophil encounter wereconsidered as core response. Core DEGs were grouped according to their transcript profile at 15 min to 60 min p.i. and plotted for yeast and hyphae.a DEGs induced in hyphae, (b) DEGs repressed in hyphae. (A + B) Genes are called commonly repressed or induced in both morphotypes if in at leastone time point in one morphotype a significantly differential expression was detected (≥ 4-fold) and in the other morphotype a regulation of ≥ 2-foldcompared to reference was given at any time point (a). Genes are called induced or repressed in hyphae but not in yeast if in at least one time pointin hyphae a significantly differential expression (≥ 4-fold) was detected and in yeast no regulation ≥ 2-fold was given at any time point (b). Genes arecalled induced or repressed in hyphae and repressed or induced in yeast if in at least one time point in hyphae a significantly differential expression(≥ 4-fold) was detected and in yeast significantly differential expression (≥ 4-fold) was given at any time point (c). c GO enrichment analysis ofcommonly (a) and not commonly induced or repressed genes (b and c). Corresponding sets of up- and down-regulated genes were mapped withthe “GO term finder” at CGD (http://www.candidagenome.org/cgi-bin/GO/goTermFinder) to biological processes. Beside the assessment of thestatistical significance of the gene set with a calculated p-value, the X-fold enrichment is calculated as ratio of percentages of the cluster frequency oftested gene set and the cluster frequency of genomic background. Red indicates up-regulation, blue indicates down-regulation. Samples from twoindependent experiments were analyzed, n = 2

Niemiec et al. BMC Genomics (2017) 18:696 Page 10 of 21

enzymes functioning in oxidation-reduction have beenboth, up- and down-regulated. Yet, the oxidation-reduction DEGs do not overlap between the twomodes of regulation. This points towards oxidativestress and a major remodeling of the metabolic pro-file of the Candida cells.The majority of genes belonging to the aforemen-

tioned metabolic pathways, induced or repressed, areregulated by the transcription factors Efg1p, Tup1p,Cap1p, and Hap43p (Fig. 6a and c) in C. albicans. WhileEfg1p and Tup1p are regulators mainly involved in mor-phologic transition, Cap1p is a dedicated regulator ofoxidative stress response [51, 52]. Hap43p was shown tobe essential for iron homeostasis in C. albicans duringinfection [53, 54].Since many DEGs identified in this study are regulated

by Hap43p, we tested whether Hap43p-deficient strainswere more susceptible to neutrophil killing than wild-type strains. For this purpose we infected neutrophilswith C. albicans SC5314 wild-type strain, a ΔHAP43mutant strain, and a complemented strain derived fromΔHAP43. Candida viability was then quantified using anATP-quantification assay. Indeed, the ΔHAP43 mutantwas significantly more susceptible to neutrophil-mediated killing than wild-type and complementedstrains. Killing nearly doubled from 43% to 75% (Fig. 6b)confirming that gene regulation governed by Hap43p isindeed essential for C. albicans survival upon neutrophilencounter.Cluster-wise analysis revealed that most pathways are

commonly induced or repressed, whereas certain path-ways were differentially regulated in yeast and hyphalcells challenged by neutrophils (Fig. 5a and b). In yeast,induction of the arginine and sulfur metabolism oc-curred, indicating high levels of oxidative stress, and re-arrangement of virulence factors (Fig. 5b). Remarkably,the arginine metabolism was oppositionally regulated inyeast and hyphae. While ARG1 and ARG3 were inducedin yeasts and repressed in hyphae, CAR1 and CAR2 wereinduced in hyphae and repressed in yeasts (Fig. 5c andAdditional file 10: Table ST5). In agreement, argininebiosynthesis, more precisely the genes ARG1 and ARG3,were shown earlier to be induced in macrophage-engulfed Candida upon oxidative burst [55].

ROS detoxifying enzyme genes are induced in yeast butnot in hyphae upon neutrophil challengeC. albicans fights ROS-mediated attacks by neutrophilswith a variety of detoxifying enzymes. Of the 6 super-oxide dismutases, Sod4p, Sod5p, and Sod6p are associ-ated to the cell surface and Sod4p and Sod5p are crucialduring interaction with phagocytes [50, 56]. Additionally,C. albicans has a catalase Cat1p and small redoxproteins like thioredoxin. Remarkably, none of those

genes was part of the ‘Candida core response’. A cut-offof 4-fold and a focus on DEGs in hyphae (‘Candida coreresponse’) led to exclusion of the majority of ROS-detoxifying proteins. Nevertheless, SOD3 and SOD4were induced more than 4-fold in yeast encountering in-tact neutrophils at one or more time points, but not inthe hypha infection (Table 4). According to direct com-parison of expression levels (by RPKM), hyphal cellsalready show high transcript levels before incubationwith neutrophils (Additional file 11: Table ST6). Thissuggests that hyphal cells might already be better pre-pared for ROS-mediated stress. In summary, a numberof Candida genes typically connected to virulence wasindeed induced under cell-culture conditions in yeasts,but was concealed in hyphae, as most of the DEGs werealready induced upon germination of hyphae.

NETs trigger similar transcriptional responses in Candidayeasts and hyphae and induce a broad stress responseIn addition to analyzing the C. albicans response to-wards intact and viable neutrophils, we added yeasts orhyphae to NETs. Both morphotypes are known to besusceptible towards NETs [15]. For this in vitro set-up,NETs were generated from neutrophils using the abioticstimulus phorbol myristate acetate (PMA). With this ap-proach we ensured having nearly complete conversion ofneutrophils to NETs [13]. Since NETs are a less dynamicstimulus than live neutrophils, we focused on one timepoint to compare to resting neutrophils as reference:30 min p.i.. When Candida yeasts were added to NETs,transcript levels of 89 fungal genes were altered. Lessgenes - a total 61 DEGs - were affected in hyphae addedto NETs (Additional file 3: Figure S2C, Additional file 8:Table ST4). Of the union of 116 DEGs, we identified 44induced and 72 repressed DEGs which we consideredfor further analysis.The transcriptional response of C. albicans upon NET

exposure indicated possible evasion mechanisms (Fig. 7aand b). Two cellular processes, sulfur amino acid metab-olism and monosaccharide transport, were identified tobe significantly up-regulated in Candida triggered byNETs, whereas fatty acid metabolism and ribosome bio-genesis, were significantly repressed (Fig. 7c). Thus, thisdata points towards a solid stress response includingnutrient uptake, oxidative stress detoxification, andstrongly repressed protein expression.Most DEGs were NET-specifically induced, with few

exceptions. When comparing the ‘Candida core re-sponse’ of 135 DEGs (Fig. 5a and b) with the responsetowards NETs comprised of 116 genes (Fig. 7a and b),we found an overlap of only 24 shared genes. Similarlyas in contact with intact neutrophils, the regulatorsEfg1p, Tup1p, Cap1p, and Hap43p were important for

Niemiec et al. BMC Genomics (2017) 18:696 Page 11 of 21

the Candida response as indicated by considerable regu-lation of a number of their gene targets (Fig. 6d).

NET-Challenged C. albicans yeast induce sugarmetabolism, hyphae induce arginine metabolismTo investigate a possible differential response of yeastand hyphae to NET attacks, we included both morpho-types of C. albicans. We found that the majority of genes

to be affected in a similar fashion. Only a small propor-tion of the NET-induced DEGs genes were differentiallyregulated encountering NETs, namely 9 genes in yeastsand 8 genes in hyphae (Fig. 6e, Fig. 7a and b). Of note,NET challenge induced arginine metabolism (ARG1 andARG3) in hyphae, but not in yeast (Fig. 7a, Add-itional file 10: Table ST5). The opposite was found forthe response to intact neutrophils, in which C. albicans

Fig. 6 Transcriptional regulators controlling C. albicans DEGs during neutrophil/NET infection. DEGs of C. albicans during encounter of neutrophilsor NET challenge were classified and visualized according to their control by transcriptional regulators, focusing on Efg1p (blue), Tup1p (red), Hap43p(green), and Cap1p (orange). a Candida genes from the GO enriched processes during neutrophil infection shown in Fig. 5c which were commonlyinduced or repressed; 28 induced genes in upper group, 16 repressed genes in lower group. b Percentage killing of C. albicans challenged for 1 hwith human neutrophils (MOI 1). Viability of C. albicans was determined using a luciferase-based ATP assay. Shown is a representative experiment intriplicate from 2 experiments in total with very similar results, n = 2 (3). Statistical analysis was performed using a One-way ANOVA with Dunett’spost-test with ** = p < 0.01 and **** = p < 0.0001. c All morphology-dependently expressed Candida genes during neutrophil infection shown in inFig. 5a and b; six hypha-induced genes in upper group, 13 hypha-repressed genes in lower group. d Candida genes from the GO enriched processesduring NET challenge shown in Fig. 7c which were commonly induced or repressed; six induced genes in upper group, 43 repressed genes in lowergroup. e All morphology-dependently expressed Candida genes during NET challenge shown in Fig. 7a and b; four hypha-induced genes in upperleft group, four hypha-repressed genes in lower left group, seven yeast-induced genes in upper right group, two yeast-repressed genes in lowerright group

Niemiec et al. BMC Genomics (2017) 18:696 Page 12 of 21

yeast induced arginine metabolism whereas hyphae re-pressed it (Fig. 5b). Further, primary sugar metabolismwas induced in yeasts (Fig. 7a).

DiscussionNeutrophils are crucial for the defense against microbialinvaders [1]. Although neutrophils have a short life spantheir life style is rather complex. In dependence of themilieu and the stimulus neutrophils launch a variety ofdifferent immune responses ranging from degranulationto phagocytosis and NET release. Notably, both C. albi-cans morphotypes are susceptible to neutrophils andNETs [15]. In the present study we therefore investigatedthe early interplay of human neutrophils with C. albi-cans yeasts and hyphae as well as responses of Candidayeasts and hyphae towards NETs by dual RNA-Seq.The number of studies analyzing transcriptional re-

sponses in neutrophils is rather limited. The few studiesconducted with different cytokines and microbes dem-onstrated that neutrophils react towards different stimuliwith a distinct transcriptional program and a core generesponse independent of the stimulus. A microarray-based comparison of neutrophil transcriptomes fromLPS-, fMLF-, and E. coli-stimulated cells identified a mu-tual response that contained a high proportion of tran-scriptional regulators [32]. Overall, most of the DEGs inthis study were assigned to transcriptional regulationand gene expression as well as signaling. Similarly, acore response and a stimulus-specific response werefound when neutrophils were stimulated, ‘primed’, withTNF-α and GM-CSF and their transcriptomes were ana-lyzed by RNA-Seq [34]. A comparable trend is reflected

in our data. When analyzing the 318 DEGs in neutro-phils encountering yeast and hypha C. albicans, we de-tected 36 hyphae-specific DEGs, thus 11% of the wholetranscriptome were differentially expressed in the pres-ence of this one stimulus. The experimental conditionschosen are known to induce C. albicans filamentationwithin a few hours, but are necessary to comfort neutro-phils. Despite the short incubation time and minimizedfilamentation, we did not find any yeast-specific DEGs.To find broader implications for signaling pathways

engaged in neutrophils encountering C. albicans we per-formed a gene set enrichment analysis against publiclyavailable expression array datasets. We revealed asignificant over-representation of up-regulated genesassociated with lipopolysaccharide-treated monocytes[57], from which the NOD-like pathway showed themost dominant overlap with our data. This suggests thatsimilarly to monocytes during antibacterial responses,neutrophils trigger NOD-like intracellular receptor path-ways upon recognition of fungal pathogen C. albicans.Of note, monocyte contamination of our neutrophilpreparations were below 0.1% as determined by flow cy-tometry analysis, virtually excluding the possibility thatsimilarities to monocyte expression profile stem fromdirect contaminations. Interestingly, NOD-like receptorsare only present intracellularly where they detect mi-crobes by their PAMPs or damage-associated molecularpatterns (DAMPs). We detected the gene NLRP3 to beinduced similarly in the yeast and in the hypha responseafter 60 min: 3.6-fold and 3.4-fold, respectively. The onlyNOD-like receptor known to detect fungi is NLRP3,which leads to inflammasome activation [48]. To

Table 4 Transcription profile of selected C. albicans effectors genes upon neutrophil challenge

C. albicans yeast [log2] C. albicans hyphae [log2]

15 min 30 min 60 min NETs 15 min 30 min 60 min NETs

SOD1 −0.38 −0.02 −1.39 −0.41 1.15 0.72 0.24 0.94

SOD2 −0.60 −0.61 −0.62 0.12 −0.40 −0.66 −0.48 0.63

SOD3 1.75 −0.68 2.98 0.55 −0.24 0.30 1.19 1.04

SOD4 1.18 2.24 2.53 0.57 0.65 1.26 1.49 0.31

SOD5 1.45 0.90 0.43 1.64 0.91 0.83 0.56 0.15

SOD6 1.45 0.90 0.43 1.64 0.84 0.42 0.41 1.31

CAT1 0.67 −0.08 −1.08 1.33 1.00 −0.04 −0.87 1.20

TRX1 0.45 0.06 −0.97 0.56 1.19 0.61 0.28 0.83

ECE1 3.81 6.63 9.22 1.15 −1.16 −1.06 −1.48 −0.22

PRA1 3.32 1.03 1.49 1.77 −0.27 0.35 0.63 0.90

ALS3 1.56 3.05 4.52 −0.37 −1.13 −1.25 −1.62 −0.52

Neutrophils Neutrophils

The transcript pattern of typical C. albicans effector genes affected during neutrophil encounter and NETs challenge is displayed by fold change. The transcriptlevels during neutrophil infection were corrected by the transcript levels in the same environment without neutrophils (medium, temperature) at every giventime point

Niemiec et al. BMC Genomics (2017) 18:696 Page 13 of 21

provide further indications for inflammasome activa-tion in neutrophils upon C. albicans infection we per-formed caspase-1 activation assays and IL-1β releasequantification using ELISA. Both assays showed in-duction upon yeast and hypha infection, with a sig-nificantly higher induction towards the pro-

inflammatory, invasive hyphal form and thus sup-ported our transcript analysis. Of note, one of theDEGs of the NOD-like pathway, RIPK2, is an estab-lished drug target for inhibiting this pathway with ex-amples of clinically approved drugs includingponatinib and regorafenib inhibiting this kinase.

Fig. 7 Transcriptional response of C. albicans challenged by NETs. Candida genes differentially expressed in yeasts and hyphae upon NETencounter were considered. DEGs were grouped according to their transcript profile at 15 min to 60 min p.i. in yeast and hypha (A + B): Genes arecalled commonly repressed or induced in both morphotypes if in at least one time point in one morphotype a significant differential expression wasdetected (≥ 4 fold) and in the other morphotype a regulation of ≥ 2-fold towards was given at any time point (a). Genes are called induced orrepressed in hyphae but not in yeast (and vice-a-versa) if in at least one time point in hyphae or yeast a significant differential expression was detected(≥ 4-fold) and no regulation ≥ 2-fold was given at any time point for the other morphotype (b and c). c Corresponding sets of up- and down-regulatedgenes were mapped with the “GO term finder” at CGD (http://www.candidagenome.org/cgi-bin/GO/goTermFinder) to biological processes. Beside theassessment of the statistical significance of the gene set with a calculated p-value, the X-fold enrichment is calculated as ratio of percentages of thecluster frequency of tested gene set and the cluster frequency of genomic background. Red indicates up-regulation, blue indicates down-regulation.Samples from two independent experiments using different blood donors were analyzed, n = 2

Niemiec et al. BMC Genomics (2017) 18:696 Page 14 of 21

Moreover, our study supports the potential of neutro-phils to be part of the cytokine-mediated orchestration atthe C. albicans infection site. In agreement with earlierstudies, we identified the cytokine genes IL1A, IL1B, andIL8 to be induced in neutrophils after 1 h of stimulationwith C. albicans, while the genes TNF, IFNG, CSF2, andIL10 were not differentially regulated. Nevertheless, wedetected the cytokine TNF-α in the supernatant of neutro-phils after 18 h of coincubation with dead C. albicans in-dicating a delayed expression [33]. This finding wasrecently supported by Duggan et al. who also revealedTNF-α in the neutrophil supernatant after 4 h of coculti-vation with live C. albicans and confirmed the secretion ofMIP-1β, IL-8, as well as the lack of Gro-α and MIF in theneutrophil supernatants [58].In contrast to transcriptional analyses in microbe-

triggered neutrophils, there are a number of microarray-based studies investigating the transcriptional responseof C. albicans to phagocytes such as neutrophils andmacrophages/monocytes. The finding that C. albicansengulfed by macrophages alters its primary metabolismas a result of starvation by inducing the glyoxylate cycle[38], was followed by studies revealing that the meta-bolic response is accompanied by dramatic reprogram-ming on many levels: For instance, the induction ofhyphal growth to enable the escape from macrophages,the activation of an oxidative stress response, DNAdamage repair or arginine biosynthesis [37]. C. albicansengulfed by neutrophils was demonstrated to stronglyinduce amino acid metabolism [59]. Interestingly, thisinduction was dependent on the presence of serum andin addition stronger when serum was fresh and notheat-inactivated. This enhancement was most probablydue to an increased phagocytosis rate. Simultaneously,studies with human blood as well as enriched neutro-phils confirmed the important role of the glyoxylatecycle and the amino acid metabolism [35, 36]. A phago-cyte can therefore be described as an environment richin amino acids and poor in glucose [55]. In our study,we focused primarily on those genes that were differen-tially expressed in hyphae. In doing so, we chose to notdiscuss a number of genes that were shown to bestrongly affected upon phagocyte encounter in previousstudies. All aforementioned studies used C. albicansyeast with MOIs of 1–5 yeast cells per phagocyte to startthe interaction between the fungus and the host cell. Inour study, we used an MOI of 1 for yeast and a similaramount of hyphae quantified by dry mass correlation.Even though we did not include serum in our experi-mental set-up and only considered DEGs with an alteredtranscript level beyond 4-fold, there were striking simi-larities to earlier studies. For instance, ICL1, known tobe induced upon challenge with blood, enriched neutro-phils, macrophages, and upon engulfment by

neutrophils, was also up-regulated in yeast encounteringneutrophils in our experiment [35, 36, 38, 50]. Remark-ably, ICL1 was similarly up-regulated in yeast challengedby NETs, even though these trap-like structures cannotentirely enclose a fungal cell. Genes from the amino acidmetabolism that were demonstrated to be induced inneutrophil-encountering yeast, e.g. MET1, MET3, ARG1,ARG3, LEU1, LEU2, and LEU4 were also induced in ourstudy [36, 59]. Interestingly, we revealed ARG1 andARG3 representing two out of 4 genes to be regulated inopposite manner in the yeast and the hyphal response,i.e. they were down-regulated in hyphae-challenged withneutrophils, but up-regulated in yeast. The transcript ofthose two genes is described to be induced in C.albicans yeast upon phagocytosis and related to ROS[55, 59]. To our surprise, both genes were induced inNET-challenged Candida hyphae, but not in yeast. Sincewe could not detect high levels of ROS in our NETpreparations (Additional file 12: Figure S6), this mightpoint to the increased accumulation of intrinsic C.albicans oxidative stress in hyphae when challenged byNETs compared to intact neutrophils.As main transcriptional regulators of the affected

metabolic pathways in C. albicans encountering neutro-phils we identified Efg1p, Tup1p, Cap1p, and Hap43p. Amutant strain deficient in Hap43p was significantly moresusceptible to killing by neutrophils and thereforedemonstrates that regulation of metabolic pathways inC. albicans by Hap43p was crucial for evasion ofneutrophil-mediated killing.In contrast to a number of earlier similar studies, we

did not include serum. Certainly, antibody-mediated rec-ognition and complement are important during the in-nate immune response against C. albicans. On the otherhand, the composition of serum, e.g. the concentrationof bacterial peptidoglycans, and therefore its impact onhyphae induction likely varies [60]. Further, we per-formed comparative killing assays in a previous study inwhich we did not see a significant difference betweenthe Candida-killing capacity of human neutrophils inRPMI with and without 2% serum [61]. This issupported by a recent study demonstrating two distinctkilling mechanisms of C. albicans by neutrophils,depending on opsonization [62]. In conclusion, we areconvinced that an in vitro infection without serum is asimplified, but nevertheless valid approach to understandthe fundaments of the neutrophil–C. albicans interaction.To account for the donor variability we included differentdonors of blood as neutrophil sources. Taken together, theoverall similarities in the yeast response to neutrophils inour study indicate the comparability of our study withearlier studies and by this establish the ground for thenew findings regarding the hyphae response to neutrophilsand the C. albicans response to NETs.

Niemiec et al. BMC Genomics (2017) 18:696 Page 15 of 21

ConclusionsWe dissect the interaction of C. albicans yeast andhyphae with intact neutrophils as well as NETs. The neu-trophil response showed a major Candida morphotype-independent and a minor hyphae-specific response.Similarly, the Candida response towards both intact neu-trophils and NETs was majorly morphotype-independent.The Candida response to NETs emerged to be distinctfrom the response to intact neutrophils. We provide indi-cation that neutrophils considerably contribute to the or-chestration of complex immunological responses uponencounter of C. albicans by i.e. inflammasome inductionand release of numerous cytokines. C. albicans in turnsought to avoid these diverse neutrophil attacks by regulat-ing required genes using the transcription factor Hap43pas a key regulator to evade killing by neutrophils. Thisknowledge can contribute to a more specific targeting ofhyphal Candida as well as modulating the Candida-induced inflammatory response during infection.

MethodsGrowth of Candida albicans and dry mass adjustmentBefore use, C. albicans clinical isolate SC 5314 was re-cultivated from a frozen glycerol stock on YPD plates for24 h at 30 °C and inoculated to an overnight YPD mediumculture (1% yeast extract, 2% Bacto peptone and 2% glu-cose) at 30 °C with continuous shaking. To harvest yeastcells, a subculture in YPD was inoculated with OD600 0.1and incubated at identical conditions for 3 h. To harvesthyphae, a subculture was inoculated at a start OD600 0.1in RPMI1640 without phenol red (w/o PR) and incubatedat 37 °C with vigorous shaking. Yeast cells and hyphaewere harvested by centrifugation at 3000×g for 5 min in apre-warmed centrifuge and washed 3 times with pre-warmed RPMI prior to the experiments. Cell numbers forCandida yeast were calculated by OD600 – CFU correl-ation (C. albicans: OD600 1 ≙ 2.9 × 107 CFUs/ml). Usingan XTT / dry mass correlation, the dry mass of yeast andhyphae per milliliter were adjusted before infection [39].

Isolation of human neutrophilsNeutrophils were isolated from venous blood of healthyvolunteers as previously described [13]. Briefly, blood waslayered 1:1 on Histopaque-1119 and separated by centrifu-gation for 20 min at 800×g. Granulocyte-rich buffy coatswere collected and cells washed with PBS with 0.5% (w/V)human serum albumin (HSA). Cells were segregated on adiscontinuous phosphate-buffered Percoll gradient: 65%,70%, 75%, 80%, and 85% by centrifugation for 20 min at800 x g. The cell layer formed at the 80% to 75% interfacewas collected and washed as before. Prior to use,neutrophils were counted, assessed for viability with try-pan blue staining and diluted to desired concentration in

RPMI1640 without phenol red (w/o PR). All assays wereperformed in RPMI1640 w/o PR, if not stated otherwise.

Flow cytometry analysis of neutrophil suspension afterisolationTo judge neutrophil purity of the isolation protocol, weanalyzed cell suspensions from 5 independent isolationswith flow cytometry. Staining was performed accordingto standard protocols with antibodies against humanCD3-FITC, CD14-APC, CD19-PE, MHC class II (HLA-DR,DP,DQ)-FITC and CD11b-APC (eBioscience or BDBiosciences). Non-specific binding was blocked by15 min pre-incubation in PBS with 2%(V/V) fetal calfserum (FCS). Peripheral blood monocytes (PBMCs) wereused as positive staining controls. Data was recorded ona FACSCantoII™ and analyzed using the FACSDiva™ soft-ware (BD Biosciences).

Preparation of NETsNeutrophils (3 × 106) were seeded into a 24-well plate.NET formation was induced as described before [13].Briefly, neutrophils were stimulated with 100 nM phor-bol myristate acetate (PMA) and incubated for 4 h at37 °C with 5% CO2. Activation and conversion ofneutrophils was controlled microscopically. After 4 h,supernatant was gently removed from NETs and 200 μlpre-warmed RPMI1640 w/o PR added.

Infection of neutrophils or NETs with Candida albicansSubcultered C. albicans was washed in pre-warmedRPMI1640 w/o PR in a pre-warmed centrifuge to allowCandida to adjust to temperature and pH. Fungal yeastand hyphae were diluted according to an XTT/dry masscorrelation to ca. Four hundred ninety microgram permilliliter [39]. Of these, 150 μl were seeded into a 24-well plate, which equals 74 μg. This dry mass corre-sponds to 3 × 106 yeast cells from a 3 h subculture inYPD and thus to MOI of 1 in a yeast infection. Neutro-phils were diluted to 1.5 × 107 cells/ml and 200 μl addedto 24-well plates to start infection. To support contact ofneutrophils and Candida cells, 24-well plates were cen-trifuged for 5 min at 300×g in a pre-warmed centrifuge.Incubation occurred for 15 min, 30 min and 60 min at37 °C with 5% CO2. An incubation of 30 to 60 min wasshown before to allow a substantial proportion ofneutrophils to interact with and engulf C. albicans yeast[36, 61]. NETs were infected in a similar manner, but theinfection was only allowed for 30 min. For each timepoint, C. albicans controls w/o neutrophils were in-cluded. Unstimulated neutrophils kept in RPMI1640 w/oPR served as a control. After the incubation, 24-wellplates were centrifuged for 5 min at 300 x g in a pre-warmed centrifuge. For RNA isolation supernatant wascarefully and quickly removed and discarded and 600 μl

Niemiec et al. BMC Genomics (2017) 18:696 Page 16 of 21

RLTplus including 1% (V/V) β-mercaptoethanol (β-ME)were added, and cell residues were recovered from wellsby vigorous scratching and repetitive pipetting. For wellsdedicated to C. albicans RNA isolation, 3.5 μl β-MEwere added to reach 1% (V/V). Similarly, cell residueswere recovered from the wells by vigorous scratchingand repetitive pipetting. The resulting suspension wasshock-frozen by dropping it into liquid nitrogen. Sus-pension ‘pearls’ were kept at −80 °C until proceedingwith the RNA isolation. All infections were performed induplicates to facilitate the parallel isolation of RNA fromC. albicans and neutrophils.

RNA isolationDisruption of frozen suspension pearls was carried outusing Mixer Mill MM 200 (RETSCH, Germany) with 30/sshaking frequency under cryogenic conditions. The pow-der was resuspended in lysis buffer RLTplus (QIAGEN,Germany), supplemented with 1% (V/V) of β-ME. Extrac-tion of total RNA was performed according to QIAGEN’sMechanical Disruption Protocol for total RNA isolationfrom yeast, using the RNeasy Plus Mini Kit. The experi-ments were performed in duplicates. RNA quality andquantity was evaluated using an Agilent 2100 Bioanalyzerwith RNA 6000 Nano or Pico Chips.

cDNA library preparation for Illumina sequencingcDNA libraries were prepared with 200 ng of total RNAand Illumina’s TruSeq RNA Sample Preparation v2protocol. Sequencing was performed using an IlluminaHiSeq2000. To assess concentration and ensure appro-priate size distribution (between 200 and 570 bp) cDNAlibraries were checked using Bioanalyzer DNA 1000chips. Sequencing run was carried out with single-end50 bp long reads following manufacturer’s instructionswith average sequencing depth of 35 and 12 millionreads for neutrophils and for Candida transcriptome,respectively.

Mapping and quantification of RNA-Seq data and DEGidentificationReads were mapped against the human reference genome[hg19] and C. albicans reference genome Assembly 21using NextGenMap (v. 0.4.12) with default settings [63]. Asummary of all analyzed samples for RNA-Seq includingsequencing depth and mapping statistics is listed inAdditional file 2: Table ST1. Downstream quantification(rpkmforgenes.py, http://sandberg.cmb.ki.se/rnaseq/) ofgenes in raw read counts as well as RPKM (= reads perkilobase of exon model per million mapped reads accord-ing to Mortazavi et al. [64]) was carried out exclusivelywith uniquely mappable reads using gencode annotationv17 and annotation provided by Grumaz et al. [65] for hu-man and for C. albicans, respectively (Additional file 11:

Table ST6). To identify differentially expressed genes(DEGs) between two conditions, we applied edgeR(version 3.4.2) with two biological replicates per conditionbased on raw read counts [66]. In order to reduce back-ground signals, we applied stringent criteria consideringonly genes with an adjusted p-value (FDR) < 0.05, a meanlog2 CPM (counts per million) > 4 and a log2 foldchange ≤ −2 or ≥ 2 for Candida genes or a log2 foldchange ≤ −1.5 or ≥ 1.5 for human genes as being signifi-cantly differentially regulated between two conditions.

Enrichment analysis of neutrophil DEGsFor the enrichment analysis, we focused on 318 DEGswhich represent genes that were differentially expressed(up- or down-regulated) with a log2 fold change of ≥ 2and a p-value of < 0.05 in at least one time point uponyeast or hyphae infection. To minimize bias inherent in asingle analysis approach we subjected this list to enrich-ment analyses using three web-based tools, ENRICHnet(http://www.enrichnet.org/), KEGG pathway analysis(https://david.ncifcrf.gov/) and GeneTrail (http://gene-trail.bioinf.uni-sb.de/). The ENRICHnet analysis wasundertaken against the STRING database (a database offunctional protein-protein interactions - https://string-db.org/) as a pre-step to a KEGG pathway analysis. Bothsteps form part of the ENRICHnet analysis pipeline andthe output file is ranked according to statistical signifi-cance as determine using a Fisher’s t test. Only one path-way was significant by this measure and this was theNOD-like receptor signaling pathway with a p value of< 0.05. By contrast KEGG pathway analysis using the DA-VID web tool (https://david.ncifcrf.gov/) simply assessesthe statistical significance of sets of genes based on over-representation between a numerically matched randombackground geneset. By these criteria the NOD-like recep-tor signaling pathway was still the most significantwith a p value of < 0.05 (again based on a Fisher’s t test)however two other pathways were also significant basedon this threshold – MAPK signaling and cytokine-cytokine receptor interaction. GeneTrail was used with amultiple testing correction (Benjamin & Hochberg 1995)to identify significant KEGG pathways based on an FDRof < 0.05. This yielded four pathways significant at < 0.05which were in ranked order of significance MAPK signal-ing, cytokine-cytokine receptor interactions, NOD-like re-ceptor signaling pathway and the hematopoietic celllineage.

ROS productionROS production was measured using a luminol-basedchemiluminescence assay. NETs from 5 × 104 neutrophilswere supplemented with 50 μM luminol (Sigma-Aldrich)and 1.2 U/ml horseradish peroxidase (Sigma-Aldrich)were seeded in a white 96-well plate (Nunc). NETs were

Niemiec et al. BMC Genomics (2017) 18:696 Page 17 of 21

infected with C. albicans at MOI 1 and 2 or left unin-fected. As reference, intact neutrophils were infected simi-larly or stimulated with 100 nM PMA (Sigma-Aldrich).The chemiluminescent signal was detected with anInfinite 200 luminometer (Tecan Nordic, Mölndal,Sweden) every 3 min for 6 h in 3 replicates. Cells werekept at 37 °C with 5 % CO2 during measurements.

Candida albicans Cultivation for bio-Plex® cytokinequantificationCandida albicans clinical isolate SC5314 was culturedovernight in YPD at 30 °C on a shaker. Cell numberswere calculated by OD600 correlation (C. albicans: 1OD600 = 2.9 × 107 cells/ml). Hyphal growth was inducedby culturing C. albicans (starting OD600 0.1) in RPMI1640) at 37 °C and yeast growth was induced in YPD at30 °C for 3 h on a shaker. Subsequently C. albicans hy-phae and yeast were killed by adding 1% (w/V) thimero-sal (Sigma-Aldrich) and incubating protected from lightovernight on a roll shaker. Cells were washed 3 times in50 ml PBS prior to cytokine experiments. A platingcontrol of washed, dead C. albicans hyphae and yeastrevealed no viable colonies after incubating for 3 d at30 °C.

Stimulation and suspension cytokine array analysisNeutrophils were seeded in sterile 96-well plates at adensity of 5 × 105 cells per well in a final volume of100 μl RPMI. Cells were stimulated with dead C.albicans hyphae or yeast at MOI 2 or left untreated for18 h at 37 °C and 5% CO2. After incubation cells werespun down, supernatants were collected and subse-quently stored at −80 °C for analysis of cytokine release.Following cytokines were measured with a Bio-Plex®

Pro human cytokine 24–Plex (Group II) and 27–Plex(Group I; both Bio-Rad, Hercules, USA): We selectedthe following cytokines from group I: IL-1β, IL-1ra, IL-8,MIP-1β, TNF-α and group II: IL-1α, IL-18, GRO-α,M-CSF, MIF; The Bio-Plex® analysis was performedusing a Bio-Plex® 200 System with Bio-Plex Manager4.1.1 (Bio-Rad) software. We compared cytokine levels insupernatants of hypha-stimulated and yeast-stimulatedneutrophils for Gro α, IL-1α, IL-1β, IL-1ra, IL-8, IL-18,M-CSF, MIF, MIP-1β and TNF α. One way ANOVA ana-lyses with Bonferroni’s posttest were performed usingGraphpad Prism Software 5.03. Samples were consideredbeing significantly different with p < 0.05.

Caspase-1 assay to quantify inflammasome activationActivation of caspase-1 was investigated using a FAMFLICA™ Caspase-1 Assay Kit (Immunochemistry,Bloomington, IN, USA). Neutrophils were seeded at1 × 105 cells per well in 96-well microplates and infectedwith C. albicans at MOI 1 for 1 h and 3 h. Afterwards,

cells were labeled with FLICA (FAM-YVAD-fmk) ac-cording to manufacturer’s instructions. As positive con-trol served neutrophils + LPS (1 μg/ml), followed byaddition of ATP (5 mM, 1 h before adding the FLICA re-agent). Cells were analyzed in a plate reader (FLUOstaromega, BMG labtech, Ortenberg, Germany) to measurecaspase-1 activity.

Quantification of IL-1β in supernatants of stimulatedneutrophilsNeutrophils were seeded at 1 × 105 cells per well in 96-well microplates and infected with C. albicans at MOI 1for 3 h and 6 h in quadruplicate. Cell-free culture super-natants were harvested and IL-1β concentration wasmeasured by ELISA (Biolegend-ELISA MAX™ 9727Pacific Heights Blvd, San Diego, CA, USA), according tomanufacturer’s instructions. As positive control servedneutrophils + LPS (1 μg/ml). Sensitivity or minimumdetectable concentration of IL-1β was 0.5 pg/ml. For theassay, a standard recombinant cytokine preparation wasused to estimate cytokine concentrations in samples.

C. albicans cell viability assayC. albicans WT SC5314, CaHap43D2–1–8-1, CaHap43R3–1–5-2, CaHap43D1–1–9-1 strains were used [53]. Viabilitywas assessed by ATP quantification using luminescentCellTiter-Glo Promega kit. The fungal strains were(1 × 105 cells/well) were challenged with neutrophils(1 × 105 cells/well) at MOI 1. The plates were incubatedat 37 °C for 1 h. Neutrophils were lyzed with Triton X-100. After 15 min incubation plates were washed withPBS. Luminescence signals were recorded using a TecanInfinite F200 microplate reader.

Additional files

Additional file 1: Figure S1. Purity analysis of neutrophil isolation.Cellular composition after Percoll gradient purification (A). Neutrophils(CD11b+MHCII−): 91.8%, monocytes (CD14+): 0%, T cells (CD3+): 0.7%, DC(CD11c+): 0.1%, B cells (CD19+): 0%, other cells (predominantly FSCint

SSCHi eosinophils): 7.4%. Three multi-color staining panels were used:CD3-FITC, CD19-PE and CD14-APC to distinguish T- and B-cells as well asmonocytes; HLA-DP/DQ/DR-FITC and CD11b-Pacific Blue™ for neutrophilsand CD11b-Pacific Blue™, CD11c-FITC for DCs. FACS blots of gatingstrategies used (B). Results from one representative of 5 independentexperiments shown. (TIFF 1323 kb)

Additional file 2: Table ST1. Mapping results. Mapping statistics ofneutrophil and Candida reads using NextGenMap. (XLSX 19149 kb)

Additional file 3: Figure S2. Overview of DEGs over time. The numberof DEGs in neutrophils during C. albicans infection (A), inPMN-treated C. albicans cells (B) and in NET-treated C. albicans cells (C)over time. Red indicates up-regulation, blue indicates down-regulation.(TIFF 206 kb)

Additional file 4: Table ST2 A-C. Differential gene expression analysisof neutrophils. Comprehensive gene expression analysis table of neutrophilsinfected with yeast (A) and hyphae (B). Each condition was tested againstthe uninfected PMN control. Union of all 318 protein-coding DEGs inneutrophils (C). (PDF 52 kb)

Niemiec et al. BMC Genomics (2017) 18:696 Page 18 of 21

Additional file 5: Figure S3. Clustering of 318 DEGs in neutrophilsinfected with C. albicans. The entity of protein-coding DEGs in neutrophilsaffected during a C. albicans infection was clustered via QT-Clusteringusing Mayday based on their fold changes over the time which werez-score normalized for better visualization purposes. The cluster profilehallmarks are indicated. (TIFF 728 kb)

Additional file 6: Table ST3. Most altered DEGs in neutrophils infectedwith C. albicans. Up- and down-regulated DEGs of neutrophils infectedwith C. albicans yeast and hyphae were sorted by their respective foldchange of expression. Positive numbers indicate the ranking amongstup-regulated DEGs; negative numbers indicate the ranking amongst thedown-regulated numbers (XLSX 4327 kb)

Additional file 7: Figure S4. Cytokine secretion by neutrophils uponC. albicans infection. Neutrophils were analyzed for cytokine releaseupon 18 h stimulation with thiomersal-killed C. albicans hyphae or liveC. albicans (initially yeast). None of the analyzed cytokines showed astatistically significant difference between stimulation with dead hyphaeor live C. albicans, indicating that dead hyphae evoke similar responses inneutrophils. Statistical analysis was performed by using a One-wayANOVA with Bonferroni’s post-test (n = 5). (TIFF 724 kb)

Additional file 8: Table ST4 A-C. Differential gene expression analysisof Candida. Comprehensive gene expression analysis table of neutrophil-orNET-treated Candida cells in yeast form (A) and in hyphae form (B). Respectivelyto the form, each condition was tested against the unchallenged, butadjusted fungal cell controls. Union of all 797 DEGs in Candida cells (C).(PDF 22 kb)

Additional file 9: Figure S5. Overlaps of DEGs in yeast and hypha C.albicans challenged with neutrophils. Overlaps of DEGs in C. albicans (A)yeast and (B) hyphae challenged with neutrophils throughout the timecourse. Overlap of morphotype-specific Candida response of (C) inducedand (D) repressed DEGs. Samples from two independent experimentsusing different blood donors were analyzed, n = 2. (TIFF 353 kb)

Additional file 10: Table ST5. Differential regulation of argininemetabolism genes in C. albicans. DEGs involved in arginine metabolismof yeast and hypha C. albicans infecting neutrophils and NETs aredisplayed. The transcript level is indicated by fold change (log2).(XLS 41 kb)

Additional file 11: Table ST6 A-D. Quantification results. Comprehensivegene quantification table of neutrophil transcriptome in read counts (A) andin RPKM (B) as well as of Candida transcriptome in read counts (C) and inRPKM (D). (XLSX 18989 kb)

Additional file 12: Figure S6. ROS levels in NET vicinity. ROS producedby in vitro released NETs and neutrophils (PMNs) were quantified by aluminol-based assay over 6 h to test for background ROS due to NETpreparation in comparison to stimulated PMNs (A + B). (A): ROS in vicinityof unstimulated, PMA-stimulated, or Candida-infected NETs; (B) ROS invicinity of unstimulated, PMA-stimulated, or Candida-infected PMNs.Averages and SD plotted of 3 replicates, n = 3. (TIFF 761 kb)

AbbreviationsATP: Adenosine triphosphate; Ca: Candida; Caspase: Cysteine aspartic acidspecific protease; CD: Cluster of differentiation; DEG: Differentially expressedgene; ELISA: Enzyme-linked immunosorbent assay; FACS: Fluorescence-activatedcell sorting; FLICA: Fluorochrome inhibitor of caspases; fMLF: formyl-Met-Leu-Phe; GI: Gastrointestinal tract; LPS: Lipopolysaccharide; ME: β-mercaptoethanol;MOI: Multiplicity of infection; NADPH: Nicotinamide adenine dinucleotidephosphate; NET: Neutrophil extracellular trap; NOD: Nuclear oligomerizationdomain; p.i.: Post infection; PAMP: Pathogen-associated molecular pattern;PBS: Phosphate-buffered saline; PMA: Phorbol-12-myristate-13-acetate;PRR: Pattern recognition receptor; RNA-Seq: RNA identification and quantitationvia next generation sequencing; ROS: Reactive oxygen species; RPKM: Reads PerKilobase Million; RPMI: Roswell Park Memorial Institute medium; TLR: Toll-likereceptor; WT: Wild type; YPD: Yeast peptone dextran

AcknowledgementsWe acknowledge Prof Chung-Yu Lan for kindly providing us the CandidaΔHAP43 knockout and corresponding revertant strain.

FundingWe would like to acknowledge funding for CFU by the Swedish ResearchCouncil VR-M 521–2014-2281, Medical Faculty Umeå University 316,886 10,Laboratory for Molecular Infection medicine Sweden (MIMS), Åke WibergFoundation M15–0108 and Kempe Foundation SMK 1453. The funders hadno role in the design of the study and collection, analysis, and interpretationof data and in writing the manuscript.

Availability of data and materialsAll data generated or analyzed during this study are included in thispublished article (and its supplementary information files). The datasets usedand analyzed during the current study are available from the correspondingauthor on reasonable request.

Authors’ contributionsConceived and designed the experiments: MJN CFU. Performed theexperiments: MJN CG DE CD MS JPL. Analyzed the data: MJN CG DE CD MSJPL IGM KS CFU. Contributed material/reagents: PS. Wrote the paper: MJNCFU. All authors read and approved the final manuscript.

Ethics approval and consent to participateBlood of healthy volunteers was obtained according to recommendations ofthe local ethical committee (Regionala etikprövningsnämnden i Umeå) asapproved in permit Dnr 09-210 M with fully informed written consent ofdonors and all investigations were conducted according to the principlesexpressed in the Declaration of Helsinki.

Consent for publicationNot applicable.

Competing interestsThe authors declare no competing interests.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

GlossaryNeutrophil core response

Entity of neutrophil genes differentially expressed at one or more timepoints upon infection with C. albicans yeast and hyphal cells.

Candida core responseEntity of C. albicans genes differentially expressed in hyphal cells uponneutrophil encounter at one or more time points.

Candida core stress responseEntity of C. albicans genes differentially expressed upon different abioticstressors [49].

Author details1Department of Clinical Microbiology, Umeå Centre for Microbial Research(UCMR) & Laboratory of Molecular Infection Medicine Sweden (MIMS), UmeåUniversity, Umea, Sweden. 2Fraunhofer Institute for Interfacial Engineeringand Biotechnology IGB, Stuttgart, Germany. 3Institute of ClinicalMicrobiology, Immunology and Hygiene, University Hospital Erlangen,Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany.4Prostate Cancer Research Group, Center of Molecular Medicine Norway(NCMM), Oslo, Norway. 5Department of Molecular Oncology, Institute ofCancer Research, Radium Hospital, Oslo, Norway. 6PCUK/Movember Centre ofExcellence for Prostate Cancer Research, Centre for Cancer Research and CellBiology (CCRCB), Queen’s University Belfast, Belfast, UK. 7University ofStuttgart IGVP, Stuttgart, Germany. 8Center for Integrative BioinformaticsVienna, Max F. Perutz Laboratories, University of Vienna, Vienna, Austria.9Present Address: Leibniz Institute for Natural Product Research and InfectionBiology - Hans Knöll Institute (HKI), Jena, Germany & Center for SepsisControl and Care (CSCC), Jena, Germany. 10Present Address: Division ofMedical Protein Chemistry, Department of Translational Medicine, LundUniversity, Malmö, Sweden. 11Present Address: The Weatherall Institute ofMolecular Medicine, University of Oxford, Oxford, UK.

Niemiec et al. BMC Genomics (2017) 18:696 Page 19 of 21

Received: 13 February 2017 Accepted: 30 August 2017

References1. Borregaard N. Neutrophils, from marrow to microbes. Immunity.

2010;33(5):657–70.2. Pillay J, den Braber I, Vrisekoop N, Kwast LM, de Boer RJ, Borghans JA,

Tesselaar K, Koenderman L. In vivo labeling with 2H2O reveals a humanneutrophil lifespan of 5.4 Days. Blood. 2010;116(4):625–7.

3. Tofts PS, Chevassut T, Cutajar M, Dowell NG, Peters AM. Doubts concerningthe recently reported human neutrophil lifespan of 5.4 Days. Blood.2011;117(22):6050–2. author reply 6053-6054

4. Netea MG, Brown GD, Kullberg BJ, Gow NA. An integrated model of therecognition of Candida Albicans by the innate immune system. Nat RevMicrobiol. 2008;6(1):67–78.

5. Netea MG, Marodi L. Innate immune mechanisms for recognition anduptake of Candida species. Trends Immunol. 2010;31(9):346–53.

6. Lindau D, Mussard J, Wagner BJ, Ribon M, Ronnefarth VM, Quettier M, Jelcic I,Boissier MC, Rammensee HG, Decker P. Primary blood neutrophils express afunctional cell surface toll-like receptor 9. Eur J Immunol. 2013;43(8):2101–13.

7. Bourgeois C, Majer O, Frohner IE, Tierney L, Kuchler K. Fungal attacks onmammalian hosts: pathogen elimination requires sensing and tasting.Curr Opin Microbiol. 2010;13(4):401–8.

8. Nusse O. Biochemistry of the phagosome: the challenge to study atransient organelle. ScientificWorldJournal. 2011;11:2364–81.

9. Nathan C. Neutrophils and immunity: challenges and opportunities.Nat Rev Immunol. 2006;6(3):173–82.

10. Urban CF, Lourido S, Zychlinsky A. How do microbes evade neutrophilkilling? Cell Microbiol. 2006;8(11):1687–96.

11. Borregaard N, Sorensen OE, Theilgaard-Monch K. Neutrophil granules: alibrary of innate immunity proteins. Trends Immunol. 2007;28(8):340–5.

12. Brinkmann V, Reichard U, Goosmann C, Fauler B, Uhlemann Y, Weiss DS,Weinrauch Y, Zychlinsky A. Neutrophil extracellular traps kill bacteria.Science. 2004;303(5663):1532–5.

13. Fuchs TA, Abed U, Goosmann C, Hurwitz R, Schulze I, Wahn V, Weinrauch Y,Brinkmann V, Zychlinsky A. Novel cell death program leads to neutrophilextracellular traps. J Cell Biol. 2007;176(2):231–41.

14. Brinkmann V, Zychlinsky A. Beneficial suicide: why neutrophils die to makeNETs. Nat Rev Microbiol. 2007;5(8):577–82.

15. Urban CF, Reichard U, Brinkmann V, Zychlinsky A. Neutrophil extracellulartraps capture and kill Candida Albicans yeast and hyphal forms. CellMicrobiol. 2006;8(4):668–76.

16. Urban CF, Ermert D, Schmid M, Abu-Abed U, Goosmann C, Nacken W,Brinkmann V, Jungblut PR, Zychlinsky A. Neutrophil extracellular trapscontain calprotectin, a cytosolic protein complex involved in host defenseagainst Candida Albicans. PLoS Pathog. 2009;5(10):e1000639.

17. Mantovani A, Cassatella MA, Costantini C, Jaillon S. Neutrophils in theactivation and regulation of innate and adaptive immunity. Nat RevImmunol. 2011;11(8):519–31.

18. Cassatella MA. The production of cytokines by polymorphonuclearneutrophils. Immunol Today. 1995;16(1):21–6.

19. Gow NA, van de Veerdonk FL, Brown AJ, Netea MG. Candida Albicansmorphogenesis and host defence: discriminating invasion fromcolonization. Nat Rev Microbiol. 2012;10(2):112–22.

20. Richardson M, Rautemaa R. How the host fights against Candida infections.Front Biosci. 2009;14:4363–75.

21. Rueping MJ, Vehreschild JJ, Cornely OA. Invasive candidiasis andcandidemia: from current opinions to future perspectives. Expert OpinInvestig Drugs. 2009;18(6):735–48.

22. Inweregbu K. Nosocomial infections. Continuing Education in Anaesthesia,Critical Care & Pain. 2005;5(I):14–7.

23. Lo HJ, Kohler JR, DiDomenico B, Loebenberg D, Cacciapuoti A, Fink GR.Nonfilamentous C. albicans mutants are avirulent. Cell. 1997;90(5):939–49.

24. Young G. The process of invasion and the persistence of Candida Albicansinjected intraperitoneally into mice. J Infect Dis. 1958;102(2):114–20.

25. Lehrer RI, Cline MJ. Interaction of Candida Albicans with human leukocytesand serum. J Bacteriol. 1969;98(3):996–1004.

26. Sohnle PG, Hahn BL, Santhanagopalan V. Inhibition of Candida Albicansgrowth by calprotectin in the absence of direct contact with the organisms.J Infect Dis. 1996;174(6):1369–72.

27. Sohnle PG, Hunter MJ, Hahn B, Chazin WJ. Zinc-reversible antimicrobialactivity of recombinant calprotectin (migration inhibitory factor-relatedproteins 8 and 14). J Infect Dis. 2000;182(4):1272–5.

28. Bianchi M, Niemiec MJ, Siler U, Urban CF, Reichenbach J. Restoration ofanti-Aspergillus defense by neutrophil extracellular traps in human chronicgranulomatous disease after gene therapy is calprotectin-dependent.J Allergy Clin Immunol. 2011;127(5):1243–52. e1247

29. Murthy AR, Lehrer RI, Harwig SS, Miyasaki KT. In vitro candidastaticproperties of the human neutrophil calprotectin complex. J Immunol.1993;151(11):6291–301.

30. Jack RM, Fearon DT. Selective synthesis of mRNA and proteins by humanperipheral blood neutrophils. J Immunol. 1988;140(12):4286–93.

31. Granelli-Piperno A, Vassalli JD, Reich E. RNA and protein synthesis inhuman peripheral blood polymorphonuclear leukocytes. J Exp Med.1979;149(1):284–9.

32. Zhang X, Kluger Y, Nakayama Y, Poddar R, Whitney C, DeTora A, Weissman SM,Newburger PE. Gene expression in mature neutrophils: early responses toinflammatory stimuli. J Leukoc Biol. 2004;75(2):358–72.

33. Fradin C, Mavor AL, Weindl G, Schaller M, Hanke K, Kaufmann SH,Mollenkopf H, Hube B. The early transcriptional response of humangranulocytes to infection with Candida Albicans is not essential for killingbut reflects cellular communications. Infect Immun. 2007;75(3):1493–501.

34. Wright HL, Thomas HB, Moots RJ, Edwards SW. RNA-seq reveals activationof both common and cytokine-specific pathways following neutrophilpriming. PLoS One. 2013;8(3):e58598.

35. Fradin C, Kretschmar M, Nichterlein T, Gaillardin C, d'Enfert C, Hube B.Stage-specific gene expression of Candida Albicans in human blood.Mol Microbiol. 2003;47(6):1523–43.

36. Fradin C, De Groot P, MacCallum D, Schaller M, Klis F, Odds FC, Hube B.Granulocytes govern the transcriptional response, morphology andproliferation of Candida Albicans in human blood. Mol Microbiol.2005;56(2):397–415.

37. Lorenz MC, Bender JA, Fink GR. Transcriptional response of CandidaAlbicans upon internalization by macrophages. Eukaryot Cell.2004;3(5):1076–87.

38. Lorenz MC, Fink GR. The glyoxylate cycle is required for fungal virulence.Nature. 2001;412(6842):83–6.

39. Hosseinzadeh A, Urban CF. Novel insight into neutrophil immune responsesby dry mass determination of Candida Albicans morphotypes. PLoS One.2013;8(10):e77993.

40. Wozniok I, Hornbach A, Schmitt C, Frosch M, Einsele H, Hube B, Loffler J,Kurzai O. Induction of ERK-kinase signalling triggers morphotype-specifickilling of Candida Albicans filaments by human neutrophils. Cell Microbiol.2008;10(3):807–20.

41. d'Ostiani CF, Del Sero G, Bacci A, Montagnoli C, Spreca A, Mencacci A,Ricciardi-Castagnoli P, Romani L. Dendritic cells discriminate between yeastsand hyphae of the fungus Candida Albicans. Implications for initiation of Thelper cell immunity in vitro and in vivo. J Exp Med. 2000;191(10):1661–74.

42. Battke F, Symons S, Nieselt K. Mayday–integrative analytics for expressiondata. BMC Bioinformatics. 2010;11:121.

43. Hess J, Angel P, Schorpp-Kistner M. AP-1 subunits: quarrel and harmonyamong siblings. J Cell Sci. 2004;117(Pt 25):5965–73.

44. Cullen EM, Brazil JC, O'Connor CM. Mature human neutrophils constitutivelyexpress the transcription factor EGR-1. Mol Immunol. 2010;47(9):1701–9.

45. Hasegawa K, Tamari M, Shao C, Shimizu M, Takahashi N, Mao XQ, Yamasaki A,Kamada F, Doi S, Fujiwara H, et al. Variations in the C3, C3a receptor,and C5 genes affect susceptibility to bronchial asthma. Hum Genet.2004;115(4):295–301.

46. Arts RJ, Joosten LA, Dinarello CA, Kullberg BJ, van der Meer JW, Netea MG.TREM-1 interaction with the LPS/TLR4 receptor complex. Eur Cytokine Netw.2011;22(1):11–4.

47. Agostini L, Martinon F, Burns K, McDermott MF, Hawkins PN, Tschopp J.NALP3 Forms an IL-1beta-processing inflammasome with increasedactivity in Muckle-wells autoinflammatory disorder. Immunity. 2004;20(3):319–25.

48. Gross O, Poeck H, Bscheider M, Dostert C, Hannesschlager N, Endres S,Hartmann G, Tardivel A, Schweighoffer E, Tybulewicz V, et al. Syk kinasesignalling couples to the Nlrp3 inflammasome for anti-fungal host defence.Nature. 2009;459(7245):433–6.

49. Enjalbert B, Smith DA, Cornell MJ, Alam I, Nicholls S, Brown AJ, Quinn J. Roleof the Hog1 stress-activated protein kinase in the global transcriptional

Niemiec et al. BMC Genomics (2017) 18:696 Page 20 of 21

response to stress in the fungal pathogen Candida Albicans. Mol Biol Cell.2006;17(2):1018–32.

50. Miramon P, Dunker C, Windecker H, Bohovych IM, Brown AJ, Kurzai O,Hube B. Cellular responses of Candida Albicans to phagocytosis and theextracellular activities of neutrophils are critical to counteract carbohydratestarvation, oxidative and nitrosative stress. PLoS One. 2012;7(12):e52850.

51. Liu H. Transcriptional control of dimorphism in Candida Albicans. Curr OpinMicrobiol. 2001;4(6):728–35.

52. Alarco AM, Raymond M. The bZip transcription factor Cap1p is involved inmultidrug resistance and oxidative stress response in Candida Albicans.J Bacteriol. 1999;181(3):700–8.

53. Hsu PC, Yang CY, Lan CY. Candida Albicans Hap43 is a repressor inducedunder low-iron conditions and is essential for iron-responsive transcriptionalregulation and virulence. Eukaryot Cell. 2011;10(2):207–25.

54. Gsaller F, Hortschansky P, Beattie SR, Klammer V, Tuppatsch K, Lechner BE,Rietzschel N, Werner ER, Vogan AA, Chung D, et al. The Janus transcriptionfactor HapX controls fungal adaptation to both iron starvation and ironexcess. EMBO J. 2014;33(19):2261–76.

55. Jimenez-Lopez C, Collette JR, Brothers KM, Shepardson KM, Cramer RA,Wheeler RT, Lorenz MC. Candida Albicans induces arginine biosyntheticgenes in response to host-derived reactive oxygen species. Eukaryot Cell.2013;12(1):91–100.

56. Frohner IE, Bourgeois C, Yatsyk K, Majer O, Kuchler K. Candida albicans cellsurface superoxide dismutases degrade host-derived reactive oxygen speciesto escape innate immune surveillance. Mol Microbiol. 2009;71(1):240–52.

57. Dower K, Ellis DK, Saraf K, Jelinsky SA, Lin LL. Innate immune responses toTREM-1 activation: overlap, divergence, and positive and negative cross-talkwith bacterial lipopolysaccharide. J Immunol. 2008;180(5):3520–34.

58. Duggan S, Essig F, Hunniger K, Mokhtari Z, Bauer L, Lehnert T, Brandes S,Hader A, Jacobsen ID, Martin R, et al. Neutrophil activation by CandidaGlabrata but not Candida Albicans promotes fungal uptake by monocytes.Cell Microbiol. 2015;17(9):1259–76.

59. Rubin-Bejerano I, Fraser I, Grisafi P, Fink GR. Phagocytosis by neutrophilsinduces an amino acid deprivation response in Saccharomyces Cerevisiaeand Candida Albicans. Proc Natl Acad Sci U S A. 2003;100(19):11007–12.

60. Wang Y, Xu XL. Bacterial peptidoglycan-derived molecules activate CandidaAlbicans hyphal growth. Commun Integr Biol. 2008;1(2):137–9.

61. Ermert D, Niemiec MJ, Rohm M, Glenthoj A, Borregaard N, Urban CF.Candida albicans escapes from mouse neutrophils. J Leukoc Biol.2013;94(2):223–36.

62. Gazendam RP, van Hamme JL, Tool AT, van Houdt M, Verkuijlen PJ, Herbst M,Liese JG, van de Veerdonk FL, Roos D, van den Berg TK, et al. Twoindependent killing mechanisms of Candida Albicans by human neutrophils:evidence from innate immunity defects. Blood. 2014;124(4):590–7.

63. Sedlazeck FJ, Rescheneder P, von Haeseler A. NextGenMap: fast andaccurate read mapping in highly polymorphic genomes. Bioinformatics.2013;29(21):2790–1.

64. Mortazavi A, Williams BA, McCue K, Schaeffer L, Wold B. Mapping andquantifying mammalian transcriptomes by RNA-Seq. Nat Methods.2008;5(7):621–8.

65. Grumaz C, Lorenz S, Stevens P, Lindemann E, Schock U, Retey J, Rupp S,Sohn K. Species and condition specific adaptation of the transcriptionallandscapes in Candida Albicans and Candida dubliniensis. BMC Genomics.2013;14:212.

66. Robinson MD, McCarthy DJ. Smyth GK: edgeR: a bioconductor package fordifferential expression analysis of digital gene expression data.Bioinformatics. 2010;26(1):139–40.

• We accept pre-submission inquiries

• Our selector tool helps you to find the most relevant journal

• We provide round the clock customer support

• Convenient online submission

• Thorough peer review

• Inclusion in PubMed and all major indexing services

• Maximum visibility for your research

Submit your manuscript atwww.biomedcentral.com/submit

Submit your next manuscript to BioMed Central and we will help you at every step:

Niemiec et al. BMC Genomics (2017) 18:696 Page 21 of 21